spf-discuss
[Top] [All Lists]

Re: [SPF v1 Draft] Last chance before I submit...

2004-10-14 18:19:07
In <53BC05C0-1C1C-11D9-B42A-000393A56BB6(_at_)glyphic(_dot_)com> Mark Lentczner 
<markl(_at_)glyphic(_dot_)com> writes:

Unless I hear gnashing of teeth, I will prepare it for submission to
the IETF as an internet-draft on Wednesday, and then begin the process
of moving it to experimental RFC status.

OK, you now hear massive gnashing of teeth.


Upon further review, I just discovered that this SPF spec says that
you are supposed to return "fail" if the domain does not exist
(RCODE 3/NXDOMAIN).  See section 4.4.


WTF!?!?!?!!!

*NEVER* should we return an SPF "fail" unless so directed by the
domain owner.  If i-hate-spf.com doesn't publish an SPF record and
wants to use a non existent domain foo.i-hate-spf.com, we should
*NOT*, *EVER*, *UNDER ANY CIRCUMSTANCES*, generate anything other
than the SPF result of "none".

DON'T YOU GUYS REMEMBER THE OBJECTIONS THAT THIS CAUSED LAST TIME IT
WAS SLIPPED INTO THE SPF SPEC????  DON'T YOU REMEMBER ALL THE PROBLEMS
WITH REJECTED EMAIL THAT IT CAUSED?  DON'T YOU REMEMBER IT BEING
REMOVED FROM THE SPEC ASAP??


HOW IN THE WORLD DID SUCH A THING GET BACK IN THE STANDARD?


Geez.


Now, I realize that Mark was not active in the SPF community when this
blow-up occurred, but Meng was.  More over, Mark mentioned in passing
that NXDOMAIN would return "fail" on #SPF in the last week or two.
When I saw it, I gnashed my teeth and screamed.  In response, Mark
assured me that he had misspoke and no such thing was in the standard.


Good grief.  Those who don't remember the past are doomed to repeat
it, and even when you remind people of the past, you are ignored.



-wayne